Function run_ssh_job 

pub async fn run_ssh_job(
    job_id: &str,
    script: &str,
    hosts: &[String],
    gpus_per_node: u32,
    framework: &str,
    backend: &str,
    args: &[String],
    log_dir: &Path,
) -> Result<i32>

Launch a distributed training job across multiple nodes via SSH.